The decode logic should be done on Cows instead of slices #336
Labels
encoding
Issues related to support of various encodings of the XML documents
enhancement
help wanted
optimization
Issues related to reducing time needed to parse XML or to memory consumption
Just like I did #330 for unescaping, the same should be done for decoding. The Cow's that are currently returned make very little sense, because they usually borrow into ephemeral data so they will always eventually have to be converted into owned data. I'm thinking of having some kind of
Bytes
-based implementation that might allow us to not have to re-allocate so often when we already have aCow::Owned
.The text was updated successfully, but these errors were encountered: